Session N37: Microbial Communities II

How do biofilms defend public goods? Mechanisms of protection against planktonic invasion

Many bacteria in nature exist in biofilms -- surface-associated colonies built around a structural polymer network produced by the embedded cells. This polymer network protects cells against physical and chemical intrusions, giving biofilm-dwelling cells a survival advantage at a significant metabolic cost. Mounting evidence shows that biofilms can also protect these public goods against invading free-swimming cells, even of the same species. However, the mechanism that biofilms use to protect themselves is unclear. Using the model biofilm forming species Vibrio cholerae -- the causative agent of the pandemic disease cholera -- we combine confocal microscopy that can resolve biofilms containing hundreds of cells at single-cell resolution with digital in-line holography, a technique that can reconstruct three-dimensional trajectories of cells swimming at speeds in excess of 100 microns per second. These methods capture interactions between biofilms and swimming cells that reveal multiple layers of defensive strategies, giving insights into bacterial ecology and disease prevention.   

Comparative Studies of Nematically Aligned Bacterial Domains in Biofilms

Pseudomonas aeruginosa and Staphylococcus aureus are biofilm forming opportunistic human pathogens which aggregate by excreting and collecting materials from the environment which protectively bind cells together. Our recent work with mucoid Pseudomonas aeruginosa grown in physiological concentrations of calcium cations revealed that calcium drives the development of protective microdomains in the biofilm. In particular, we observed large scale plates of nematically aligned bacteria dubbed “shields.” In this work, we are interested in exploring the generality of these structures across several P. aeruginosa mutant strains and S. aureus, as well as characterizing the nematic order parameter of shield structures across length scales where present. We hope that a comparative study of S. aureus, if it forms shield structures, will inform our understanding of the aggregation mechanism at play. To do this, we will perform confocal fluorescence microscopy studies to produce and analyze 3D reconstructions of biofilm architecture at various calcium cation concentrations.    

Active pH regulation facilitates biofilm development in minimally buffered environments

Biofilms provide individual bacteria many advantages, yet dense cellular proliferation can also create intrinsic metabolic challenges including excessive acidification. Because such pH stress is commonly masked in buffered laboratory media, it remains unclear how biofilms cope with minimally buffered natural environments. Here, we report Bacillus subtilis biofilms overcome this intrinsic metabolic challenge through an active pH regulation mechanism. Specifically, we find that biofilms can modulate their extracellular pH to the preferred neutrophile range, even when starting from acidic and alkaline initial conditions, while planktonic cells cannot. We associate this behavior with dynamic interplay between acetate and acetoin biosynthesis and show that this mechanism is required to buffer against biofilm acidification. Furthermore, we find that buffering-deficient biofilms exhibit dysregulated biofilm development when grown in minimally buffered conditions. Our findings reveal an active pH regulation mechanism that could be targeted to control biofilm growth.   

Morphodynamics of growing bacterial communities in polymeric environments

Many bacteria live in polymeric environments, such as mucus in the body, exopolymers in the ocean, and cell-secreted extracellular polymeric substances (EPS) that encapsulate biofilms. However, most studies of bacteria focus on cells in polymer-free fluids. How do interactions with polymers influence the behavior of bacterial communities? To address this question, we experimentally probe the growth of non-motile Escherichia coli in solutions of inert polymers. We find that, when the polymer is sufficiently concentrated, the cells grow in striking “cable-like” morphologies—in stark contrast to the compact morphologies that arise in the conventionally-studied polymer-free case. Experiments and agent-based simulations show that these unusual community morphologies arise from an interplay between polymer-induced entropic attraction between pairs of cells and their hindered ability to diffusely separate from each other in a viscous solution. These results suggest a pivotal role of polymers in regulating microbe-host interactions, and more broadly, this work helps to uncover quantitative principles governing the morphogenesis of diverse forms of growing active matter in polymeric environments.   

The effects of substrates on biofilm growth

Bacteria reproduce and generate extracellular polymeric substance (EPS) to form biofilms, multicellular communities of bacteria often found on surfaces. Biofilms are complex systems where individuals perform specific tasks based on their intracellular signaling networks and the environmental cues that they receive. Several experiments have also observed biofilms expand at rates that depend on substrate stiffness, but the exact mechanisms behind this collective behavior are not well understood. We present an agent-based model to address these observations and suggest a possible mechanism for biofilm expansion on substrates characterized by their stiffness, where the individual's tasks are coupled to nutrient availability and substrate properties. We apply our model to current and ongoing experimental studies of biofilm dynamics and growth on dynamical substrates.   

Creating multi-antibiofilm therapeutics with cellulase based carbon dots

Pseudomonas aeruginosa biofilms can cause chronic and serious wound infections. Treatment of wound biofilms is difficult, and often requires multiple rounds of antimicrobial therapy and debridement. There is a need to create topical therapeutic treatments with multiple antibiofilm effects.  In this work, we explore the potential of cellulase based carbon dots as a potential therapeutic. Cellulase is a glycoside hydrolase that has already been shown to be effective in causing biofilm dispersal as well as reduction in biofilm stiffness and elasticity.  By synthesizing the cellulase into a carbon dot, we hypothesize that the molecule can retain some of those capabilities, because carbon dots are known to retain some of the effects of their precursor proteins and increase the bactericidal effect and penetration of the cellulase.

Carbon dots are synthesized from both cellulase and citric acid using a one-step hydrothermal carbonization method, and then characterized for surface properties, size, roughness, compositions, and emission spectra.  P. aeruginosa biofilms were grown in a microfluidic environment for 24 hours and then treated with these carbon dots at a range of concentrations.  Using microrheology, it was seen that biofilms treated with the cellulase based carbon dots were more viscous and less stiff compared to those treated with citric acid-based carbon dots.  This appears to be because the cellulase carbon dots impact PSL within the biofilm based on microrheology studies examining bacteria strains with specific EPS component removed.  This is consistent with the impact of molecular cellulase on biofilms. The penetration of the carbon dots into the biofilms is being examined via confocal microscopy. In general, initial studies suggest that the cellulase based carbon dots have the potential to act as an excellent multi-antibiofilm therapeutic in the future.

    

Incorporation of collagen into bacterial biofilms impedes phagocytosis by neutrophils

Biofilms are communities of microbes embedded in a matrix of extracellular polymeric substances (EPS).  EPS can be produced by biofilm organisms and can also originate from the host. The biofilm matrix protects bacteria from clearance by the immune system, and some of that protection likely arises from the mechanical properties of the biofilm. It has been shown that collagen, a host-produced protein abundant in many infection sites, can be incorporated into biofilms and change biofilm mechanics. Here we use two biofilm-forming human pathogens, Pseudomonas aeruginosa and Staphylococcus aureus, to study the role of incorporated collagen in the immune clearance of biofilms by human neutrophils. We also investigate the effects of enzymatic breakdown of host-derived materials on biofilm mechanics and susceptibility to host immune clearance. Microrheology, SEM and reflectance confocal microscopy can characterize how incorporated collagen affects biofilm microstructures and mechanical properties.   

Climate change and crops: How high-performance modeling of bacterial-fungi communities can improve food security and biofuel production in an ever-changing climate.

Climate change threatens the makeup of our environment and will place pressure on current agricultural practices throughout the USA and beyond. This necessitates the development and implementation of techniques to optimize crop growth in current agricultural setups and to further enable crop growth in previously inhospitable regions. One promising technique is the introduction of microbial communities into crop farmland whose synergistic interactions have been shown to promote an improved plant biomass yield and resilience to environment perturbations. However, the complex interactions that occur are challenging to understand and performing bio-secure experiments in-situ is unfeasible. Therefore, advanced computational models are required to study these bacteria-fungi-crop community interactions. We develop hybrid continuum-discrete models for physics and biology informed modelling of the complex interactions between microbes and their shared environment. Our multiscale models utilize high performance GPU-accelerated techniques to model the large-scale systems in realistic detail. This software can be used to understand bacteria-fungi interactions and can suggest potential bioengineering interventions to help guide farming practices in an ever-changing climate.   

Rapid antibiotic susceptiblity test using white-light interferometry

Antimicrobial resistance (AMR) has been coined the “silent pandemic” as it is a current and growing threat to global health. AMR has caused millions of deaths worldwide and leads a more than 10% treatment failure rate. Without effective antimicrobials, modern medical advances such as transplants, chemotherapy, and treatment of premature infants may become unsafe. One approach to reduce treatment failure is to improve diagnostics, i.e., more accurately determine if antimicrobials will work. With the existence of and potential for uncharacterized resistance mechanisms, phenotypic tests are necessary for clinical diagnostics. Current gold-standard tests for antimicrobial susceptibility, however, lack the sensitivity to rapidly differentiate susceptibility phenotypes and generally discount heterogeneous population responses. The best-case scenarios even for bulk population responses require tens of bacteria doubling times.

Our previous work uses the surface topography of a community to detect phenotypically heterogeneous populations. Now we take advantage of the coffee ring effect to develop a novel method of relative cell counting using volume from white light interferometry as a proxy for number of cells. This technique will reduce the number of doubling times required for detection. Quickly differentiating resistant phenotypes will enhance clinical treatment information which will improve antibiotic stewardship as well as treatment outcomes.   

Using gnotobiotic living food to study gut bacterial stability in zebrafish

The dynamics of gut microbial communities influence the health of humans and animals. Community composition is known to fluctuate strongly, but the factors that set the relevant timescales remain minimally studied due to the difficulty of generating controllable experimental systems, especially in which the role of food ingestion can be separated from the role of microbial input. To address this, we have developed a protocol to eliminate the innate bacteria in the micro-animal rotifers and use them as a nutrient source for zebrafish larvae, a model vertebrate organism. The method consists of UV light cycles to reduce the native bacterial population, keeping a fraction of motile rotifers that fish can capture. We examine and quantify the effect of rotifer ingestion on the population of a commensal bacterium, Enterobacter (EN), finding a decline with a timescale of days. We further examine the impacts of feeding on gut communities composed of up to five bacterial species. The use of bacteria-depleted rotifers enables studies of a range of physiological properties, such as fish size, intestinal transport mechanics, and overall metabolic activity, under conditions that can separate the impacts of feeding and bacterial presence.   

Predicting metabolic response to dietary intervention using deep learning

Due to highly personalized biological and lifestyle characteristics, different individuals may have different metabolic responses to specific foods and nutrients. In particular, the gut microbiota, a collection of trillions of microorganisms living in our gastrointestinal tract, is highly personalized and plays a key role in our metabolic responses to foods and nutrients. Accurately predicting metabolic responses to dietary interventions based on individuals' gut microbial compositions holds great promise for precision nutrition. Existing prediction methods are typically limited to traditional machine learning models. Deep learning methods dedicated to such tasks are still lacking. Here we develop a new method McMLP (Metabolic response predictor using coupled Multilayer Perceptrons) to fill in this gap. We provide clear evidence that McMLP outperforms existing methods on both synthetic data generated by the microbial consumer-resource model and real data obtained from six dietary intervention studies. Furthermore, we perform sensitivity analysis of McMLP to infer the tripartite food-microbe-metabolite interactions, which are then validated using the ground-truth (or literature evidence) for synthetic (or real) data, respectively. The presented tool has the potential to inform the design of microbiota-based personalized dietary strategies to achieve precision nutrition.   

Uncovering Molecular Mechanisms of Legionnaires' Disease: Coarse-Grained Martini Simulations of T4SS

Legionnaires' disease is a serious lung disease caused by Legionella pneumophila. The type IV secretion system (T4SS) is central to the infection process by delivering effector proteins into host cells. Recent breakthroughs have determined the atomistic structure of the T4SS, however much is still unknown about the molecular mechanisms behind the its function and potential as a drug target. To develop simulations of relevant duration, we built a coarse-grained model using the Martini force field for molecular dynamics in GROMACS. To validate the thermodynamic properties of the Martini force field during our simulations, we benchmarked the performance of computed signatures for Martini simulations against fully atomistic simulations in several simple cases.   

Co-aging complex systems: Featuring infected worms, competing trees and chess battles

Aging, as defined in terms of the slope of failure probability versus time, is a generic phenomenon observed in nearly all complex systems. Interdependency networks can accurately describe the aging statistics of biological species as well as complex mechanical devices: In an interdependency network, when one component malfunctions, so do others that depend on it, causing a cascading failure. This model, together with its evolutionary counterparts, predict failure rates that strictly increase in time. However, hazard curves with peculiar ups and downs have been observed in nature in seeming contradiction with theory. Here we introduce the concept of "co-aging", where the demographics of multiple cohorts are intertwined and show that co-aging dynamics can account for these peculiar bumps empirically observed in mortality curves. Specifically, we introduce a model where multiple interdependency networks inflict damage on each other, in addition to experiencing intrinsic damage. We then successfully fit our model predictions to the experimental failure statistics for (1) co-aging worm-pathogen populations (2) multiple competing tree species, and (3) machine-against-machine chess games. Importantly, our model can successfully fit not only cumulative failure rates, but also cause-specific failure rates, e.g. distinguishing between deaths that primarily stem from the accumulation of intrinsic damage versus extrinsic assaults.   

Theoretical Modeling of Sporulation Patterns in Bacillus subtilis Biofilms

Many species of bacteria, including Bacillus subtilis, undergo a transition to a spore state. Traditionally understood as a response to stress or starvation, sporulation is suppressed in lab strains of B. subtilis under healthy conditions. However, we find that sporulation emerges naturally in wild isolates of B. subtilis under the same conditions. The two strains also form structurally different biofilms. Specifically, the lab strain forms biofilms with mostly matrix-producing cells, whereas the wild strain has matrix-producing cells at the edge of the biofilm and spores in the center. We hypothesize that sporulation in wild isolates occurs via a temporal program, and we develop a model to test whether this hypothesis explains the biofilm structure. Our model has three terms – a cell growth term, a mechanical expansion term, and a temporal sporulation term – and admits traveling wave solutions, similar to the traditional Fisher-KPP equation. The model robustly reproduces the experimentally observed pattern with cells at the edge and spores in the center, regardless of biologically motivated modifications to the three terms. It also makes predictions for how the mechanistic parameters control the biofilm expansion speed, the spore fraction, and the width of the edge-cell regime, all of which are experimentally measurable. We will discuss comparisons of these predictions to experiments that modulate sporulation rate, mechanical properties, and nutrient availability. Our model is generic and can apply to other problems in which temporal state changes give rise to spatial patterns in expanding media.